Hi folks,
I’m currently trying to repair a Rega Mira 1 Clamshell with a DC offset problem. The issue is that DC offset in the right channel is quickly rising when turning up the volume. DC Offset at zero volume is around 10mV. From the 8 o’clock setting the DC offset in the right channel starts rising. At around the 9 o’clock setting the DC offset has risen to ~300mV and the protection circuit kicks in.
What I've checked so far:
Any tips on how to troubleshoot this issue?
Thanks you.
PS. Circuit diagram for Rega Mira 2000 (similar): Rega Mira 2000
I’m currently trying to repair a Rega Mira 1 Clamshell with a DC offset problem. The issue is that DC offset in the right channel is quickly rising when turning up the volume. DC Offset at zero volume is around 10mV. From the 8 o’clock setting the DC offset in the right channel starts rising. At around the 9 o’clock setting the DC offset has risen to ~300mV and the protection circuit kicks in.
What I've checked so far:
- Power supply seems fine. Power rails +/-15V, +/-40V measure ok with low ripple. Caps have been replaced.
- Preamplifier output looks fine so it looks like a problem in the right power amp circuit.
- When injecting a test signal, the sine waveform in both power amp channels looks fine at various volume levels (no clipping).
- I’ve checked the power transistors and compared some voltage measurements on the left and right output circuits but haven’t been able to find any obvious faults.
Any tips on how to troubleshoot this issue?
Thanks you.
PS. Circuit diagram for Rega Mira 2000 (similar): Rega Mira 2000
I will try to join you in your research because I also have the same amp which seems to me to have the same problem and which has burned out 3 times in a row.
I would check/replace, C36 and C48. If they become leaky, that will cause a DC offset as will the same issues with C49 & C59.
Finally had some time to work on the Rega amp...
I checked C36/C48 and they both measured ok (these have already been recapped so did not expect these to be faulty). But when I swapped them, the problem was now in the left channel, so I was pretty convinced that I found the problem. However when replacing the left capacitor with a new one the DC offset was still there when turning up the volume. Not sure how this can be explained, maybe I initally criss-crossed the left and right speaker wires.
What I did next was turning up the volume so that the DC offset increased in the left channel and doing a number of voltage checks. The point in the circuit where measurements were different in both channels was R67. I measured ~60mV at the input of R67 but ~280mv at the output, whereas in the right channel there was no significant change.
I guess this means a problem downstream in the left power amplifier circuit? Any tips on how to proceed?
I checked C36/C48 and they both measured ok (these have already been recapped so did not expect these to be faulty). But when I swapped them, the problem was now in the left channel, so I was pretty convinced that I found the problem. However when replacing the left capacitor with a new one the DC offset was still there when turning up the volume. Not sure how this can be explained, maybe I initally criss-crossed the left and right speaker wires.
What I did next was turning up the volume so that the DC offset increased in the left channel and doing a number of voltage checks. The point in the circuit where measurements were different in both channels was R67. I measured ~60mV at the input of R67 but ~280mv at the output, whereas in the right channel there was no significant change.
I guess this means a problem downstream in the left power amplifier circuit? Any tips on how to proceed?
is it this amp model ?
https://pinkfishmedia.net/forum/threads/original-clamshell-rega-mira.171657/
https://www.canuckaudiomart.com/det...mshell-rega-mira-integrated-stereo-amplifier/
according your observations I would try at first step to disconnect the volume control input from the preamp output (only in case of using a usual potentiometer like ALPS RK27.
If no change, there are transition resistance anywhere on PCB between attenuator and voltage rails. But I think rather, that the offset problem does no longer present.
Then the preamp output provide an offset - in most cases the electrolytic caps in the signal pad between preamp out and volume control input should be replace by a MKS version - e. g. from WIMA.
If a digital volume control chip is in use and a completely DC pad, the chip is to replace (burr-Brown, Wolfson, National Semiconductor etc).
If the attached REGA MIRA 2000 circuit (with digital volume control from WOLFSON) concerning the from me mentioned parts in question is identical to your device - what is the aim of RLA6 ?
Last edited:
~60mV and ~2mV. See diagram attached.
I'll do some additional measurements at the points suggested and report back.
Yes, that is the amp model. Mira 1 clamshell with ALPS potentiometer.
Can a faulty volume potentiometer introduce a DC offset problem? I guess I could test with a test signal fed into the Preamp In RCA's and switching off the preamp circuit.
I can't imagine but if the same behaviour is to observe after disconnect the Alps potentiometer from the preamp output, there is an insulation fault somewhere in the device to the potentiometer wiring.
Please measure the voltage at the input pin L+R on the Alps potentiometer with connected and disconnected preamp output.
Ok so did a few more measurements and now I'm even more confused...
Test 1
Input signal fed into Preamp IN
Preamp-Power amp internal dip switch set to OFF
Multimeter and oscilloscope connected to speaker output
Volume slowly raised to ~10 o'clock
Result: normal sine wave on speaker output in both channels, DC offset quickly rising in RIGHT channel.
Test 2
Input signal fed into AUX1
Preamp-Power amp internal dip switch set to ON
Multimeter and oscilloscope connected to speaker output
Volume slowly raised to ~10 o'clock
Result: sine wave only visible in left channel, DC offset quickly rising in LEFT channel.
Test 3
Input signal fed into AUX1
Preamp-Power amp internal dip switch set to ON
Multimeter connected to speaker output, oscilloscope connected to Preamp OUT
Volume slowly raised to ~10 o'clock
Result: sine wave only visible in left channel, DC offset quickly rising in LEFT channel.
Does anyone have a clue how the DC offset can switch channels depending on power amplifier input?
Test 1
Input signal fed into Preamp IN
Preamp-Power amp internal dip switch set to OFF
Multimeter and oscilloscope connected to speaker output
Volume slowly raised to ~10 o'clock
Result: normal sine wave on speaker output in both channels, DC offset quickly rising in RIGHT channel.
Test 2
Input signal fed into AUX1
Preamp-Power amp internal dip switch set to ON
Multimeter and oscilloscope connected to speaker output
Volume slowly raised to ~10 o'clock
Result: sine wave only visible in left channel, DC offset quickly rising in LEFT channel.
Test 3
Input signal fed into AUX1
Preamp-Power amp internal dip switch set to ON
Multimeter connected to speaker output, oscilloscope connected to Preamp OUT
Volume slowly raised to ~10 o'clock
Result: sine wave only visible in left channel, DC offset quickly rising in LEFT channel.
Does anyone have a clue how the DC offset can switch channels depending on power amplifier input?
Last week I had another look at the Rega Mira and made some progress. I can now explain some of the weird behavior, although the DC offset problem still isn't solved.
One of the issues was that there was no output on the right channel when feeding a test signal to Aux1, indicating a problem in the preamp section. The cause for this was a number of cold solder joints around the volume pot and selector switch. After resoldering these joints, the signal output is now present in both channels.
In previous tests I noticed that the DC offset seemed to switch channels in some cases, which was rather perplexing. I finally realized what was happening... The DC offset in fact appears in both channels at a certain volume knob position but the left channel is lagging a bit behind (I tested this by disconnecting the right RCA cable the AUX1 input). Before I could see the DC offset rising in the left channel the speaker relay had already disengaged due to the high offset in the right channel. Only in test #3 I noticed the offset in the left channel as the right channel signal never made to the power amp stage.
Next, I tried to narrow down to the cause for the DC offset problem in both channels but so far without success. Some things I tried was switching TR26/29 with TR33/34 and switching TR27 with TR23. This did not seem to make a difference (high DC offset first appearing in the right channel).
Problem recap:
Does anyone have any tips on how to proceed troubleshooting? Cheers.
One of the issues was that there was no output on the right channel when feeding a test signal to Aux1, indicating a problem in the preamp section. The cause for this was a number of cold solder joints around the volume pot and selector switch. After resoldering these joints, the signal output is now present in both channels.
In previous tests I noticed that the DC offset seemed to switch channels in some cases, which was rather perplexing. I finally realized what was happening... The DC offset in fact appears in both channels at a certain volume knob position but the left channel is lagging a bit behind (I tested this by disconnecting the right RCA cable the AUX1 input). Before I could see the DC offset rising in the left channel the speaker relay had already disengaged due to the high offset in the right channel. Only in test #3 I noticed the offset in the left channel as the right channel signal never made to the power amp stage.
Next, I tried to narrow down to the cause for the DC offset problem in both channels but so far without success. Some things I tried was switching TR26/29 with TR33/34 and switching TR27 with TR23. This did not seem to make a difference (high DC offset first appearing in the right channel).
Problem recap:
- At low volume DC offset in both channels seems perfectly normal (stable around 0-10 mV)
- When raising volume, sine wave looks good on both speaker outputs (no oscillations at any volume)
- At around 10am position, DC offset in right channel quickly rises to around 300mV causing the speaker relay to disengage.
- At 11am position, DC offset in left channel starts to rise quickly as well (right channel input disconnected).
- Preamp section and volume control seems ok (high DC offset remains when feeding line level signal to power amp inputs, bypassing volume control).
Does anyone have any tips on how to proceed troubleshooting? Cheers.
You've already discovered that the soldering of the board is bad. This has likely taken years to manifest itself and can only get worse or spread further unless you systematically re-solder all PCB component joints, using plenty of flux and then cleaning up properly with IPA for example, afterwards. Before starting, I'd record (on a copy of the schematic) the DC voltages at the terminals of all semis at least, for comparison with what you find after re-soldering. When done, systematically retest all those strange, varying voltage faults and verify that all semis do have appropriate or at least similar and stable voltages at their connections in each channel. I suggest you don't try blasting sound through the amp yet - not until you establish that DC conditions at the outputs remain safe, such that you can't fry your speakers, whether the protection relay(s) is functional or not.
Last edited:
Thanks Ian, that makes sense. I guess the fact that for this amp model the motorized selector switch engages and rotates automatically on every power up cycle leads to problems with the PCB solder joints over time.
Btw, I have the speakers disconnected, I'm using a resistive load on the speaker outputs.
Btw, I have the speakers disconnected, I'm using a resistive load on the speaker outputs.
If I understand correctly, the voltages across R67 shown in post 11 are the persistent problem. Is there any DC voltage across R72? (Compare with volts across R51 for reference.) If so, offset problem is likely due to leakage in C48.
So C48 is probably not the problem, but I advise checking across R72 anyway--- might provide insight. I agree with Nigel that input input diff transistors are under suspicion. But left channel is problem, so we're talking TR33 and TR34--- correct?
I note that base current should flow out of TR33 and drop about 60mV to yield about 0V at output. Nominal current bias current would be about 2.7uA. But the measurements suggest current through R67 is about 100uA and in the wrong direction. This current has to come from somewhere. Perhaps leakage in TR33.
I note that base current should flow out of TR33 and drop about 60mV to yield about 0V at output. Nominal current bias current would be about 2.7uA. But the measurements suggest current through R67 is about 100uA and in the wrong direction. This current has to come from somewhere. Perhaps leakage in TR33.
😳 That should have been 10uA, not 100uA. But the issues remain the same.
I went though college with a slide rule. You'd think I'd be better with decimal points. 😕
I went though college with a slide rule. You'd think I'd be better with decimal points. 😕
Today I took a number of voltage measurements in the power amp circuit with a rather surprising result: when raising the volume both the positive and negative power rails are gradually being dragged down. At the ~10 o'clock volume level when the DC offset appears, the power rails are as low as 25V instead of the 42V with volume at zero. I guess there is a bigger problem than just dc offset...
Volume at zero:
Volume at 10 o'clock:
Volume at zero:
Volume at 10 o'clock: